Official Helium Community - hip-58-poc-distance-limit

## HIP 58: PoC Distance Limit - Author(s): @hashc0de , @mrpatrick1991, et al - Start Date: 2022-04-02 - Category: Technical - Original HIP PR: https://github.com/helium/HIP/pull/381 - Tracking Issue: https://github.com/helium/HIP/issues/384 - Status: In Discussion ## Summary The Proof-of-Coverage incentive model is crucial to the ongoing growth of the Helium Network. As the network grows, efforts to prevent Proof-of-Coverage gaming must be bolstered in order to encourage honest Hotspot deployment. One known gaming technique works by altering the information provided by beacon and witness packets to increase reported distances. This can make it appear that Hotspots provide coverage where they do not. An effective way to reduce the impact of this attack is to place an upper limit on the distance at which witnesses for hotspots will be considered valid by the blockchain and thus be considered for rewards. When PoCv11 was developed, a witness distance limit feature was written but never activated. If approved, this proposal would impose a 100 km limit to Proof-of-Coverage witnessing and this value can be modified further through governance in the future. Witnesses where the asserted distance is greater than 100 km would be marked invalid by the Challenger. This limit is further supported by the fact that most sensor deployments have a practical maximum distance of 30-50km. Although a Helium Hotspot may be able to see PoC beacons greater than 100 km, there are no current use cases at this range. # Rendered View https://github.com/helium/HIP/blob/main/0058-poc-distance-limit.md

HIP 58: PoC Distance Limit · Issue #384 · helium/HIP

HIP 58: PoC Distance Limit Author(s): @abhay, @mrpatrick1991, et al Start Date: 2022-04-02 Category: Technical Original HIP PR: #381 Tracking Issue: this Status: In Discussion Summary The Proof-of-...

HIP/0058-poc-distance-limit.md at main · helium/HIP

Helium Improvement Proposals. Contribute to helium/HIP development by creating an account on GitHub.

edaatweets pinned a message to this channel. 04/04/2022 9:59 AM

100km sounds more than fair (although I do get the occasional 150 to 200km witnesses due to the flatness of the terrain..) (edited)

10:07

When can we vote?

obligatory tl;dr post PoC activity should continue to reward well placed hotspots. The current PoC mechanics allow manipulation of witness RF data at long distances and thus, we find evidence of gaming at the edges. This improvement implements a sanity check filter on Proof-of-Coverage and reduces the maximum range of PoC activity to 100km. We arrived at this number by considering that typical devices are able to be heard 30-50km away. We doubled that number to incentivize new Hotspot placements. The impact on a few legitimate witness events that are > 100km is limited as described in the dataset in the HIP. Please read up on the HIP to learn more. Open to questions here.

2

>100km witnesses are really cool to have, but don't realistically add anything to the network

While I agree with the 100k limit and support it, 10k witness events is a relatively small sample size, that's 3 to 4 blocks worth of witnesses.

Keep reading, there's a 4k block sample later

150kms/200km or more would be better at least the hip wouldn't affect those who have got a got mountain location, and would only affect spoofers, as long as it can really transmit that distance why would we handicap the system just because they are currently 'no current case uses' is a terrible way to limit innovation long range applications in the future.

1

MoreDPS

150kms/200km or more would be better at least the hip wouldn't affect those who have got a got mountain location, and would only affect spoofers, as long as it can really transmit that distance why would we handicap the system just because they are currently 'no current case uses' is a terrible way to limit innovation long range applications in the future.

if there are long range applications in the future, then certainly we can expand the distance limit to 150km or 200km. the nice thing is that this is a chain var. we can use governance to change it the same way we're imposing it today.

1

I live in the flattest region of Italy and i approve of this implementation

I'll lose a few witnesses, but if it helps with combating cheaters....go for it! (edited)

hashc0de

Keep reading, there's a 4k block sample later

I read that, was more referring to estimating a distribution based on a relatively small sample. I ran some queries against the ETL and arrived at more or less the same values so I'll shut up now

As I've said before on this topic, great way to better align PoC with network usage

2

1

"Witnesses where the asserted distance is greater than 100 km would be marked invalid by the Challenger."

11:20

Why mark them invalid?

continues to give us information. similar to marking witness too close as invalid. (edited)

Especially in urban scenarios there are >14 witnesses per beacon

11:21

Just ignore the one over 100km and pick a closer hotspot that also received the beacon

i guess this wasn't clear, btw. this current implementation is something that can be turned on days after a vote is completed. code is already on chain and has been since november.

1

FezzFest

Just ignore the one over 100km and pick a closer hotspot that also received the beacon

That kills all possibility for potential analysis on distances in the future.

1

You're also preventing spoofers from causing denial of service attacks this way

groot

That kills all possibility for potential analysis on distances in the future.

this as well. if we do want to expand the distance in the future, we'd have the data available.

FezzFest

You're also preventing spoofers from causing denial of service attacks this way

They will mostly be making the remaining people happy because they will have well rewarded witness events. (edited)

Yeah sure, a beacon with 14 invalid witnesses doesn't make the spoofers any money, but it also f*cks over the rewards for every legit hotspot in the vicinity

groot

They will mostly be making the remaining people happy because they will have well rewarded witness events. (edited)

Assuming the area is not dominated by spoofers

11:26

And spoofers "coverage areas" can get pretty big, as we've seen in the past

Not really relevant here, but given enough spoofers that 5 of 14 places remain. The legit hotspots will share equal odds for those 5 spots. In the long run the legit hotspots will have an equal share of high-reward witness events within those 5 spots. Less consistent rewards, but similar to the whole 25, 18 or 14 witness beacons, it should work out the same.

groot

That kills all possibility for potential analysis on distances in the future.

Why? It's not because you select another witness from the pool you can't keep a record of the invalid witnesses

groot

Not really relevant here, but given enough spoofers that 5 of 14 places remain. The legit hotspots will share equal odds for those 5 spots. In the long run the legit hotspots will have an equal share of high-reward witness events within those 5 spots. Less consistent rewards, but similar to the whole 25, 18 or 14 witness beacons, it should work out the same.

Sure, unless you have a pool of >300 spoofers and 5 legit hotspots. Chances any of those 5 legit hotspots are selected to receive the beacon is rather low (edited)

11:30

The beacon will simply have 14 invalid witnesses and no one gets paid

considering the end goal = data transfer. I think this makes perfect sense.

️

1

Make it 150 km i think that would be perfect for us all, i live on a hill and some of my witnesses are over 130km, i think 150 would be perfect.

Make it 8.3km. None of my witnesses are further away than that.

4

backtran

Make it 8.3km. None of my witnesses are further away than that.

I’m sure we can bikeshed the number forever. I’d rather move to a vote if the proposal and the way the number was reached makes sense to people. This is weird one because: 1. Code is already written and ready to activate. 2. This vote would be both on the framework (a distance limit at all) and a value (100km) 3. We also assume that future proposals can change the number. Moving to a vote after 3-4 days and having a week of voting makes sense to me.

5

Fizzy pinned a message to this channel. 04/04/2022 3:14 PM

Fizzy pinned a message to this channel. 04/04/2022 3:15 PM

This hip is big stupid yo lmao

15:41

My offgrids send more data than any residential miners

15:41

They need the far witness to make them financially sound

15:42

And they cover large areas of uncovered land

15:42

So this hip would in essence kill offgrids which provide very large areas of good lorawan coverage

At the current limit of 100km I formally dissent against this hip

100km is a good target for the reasons stated above. Double the realistic data transfer range (the whole point of the network), and yet still gives PoC rewards to help grow. A 100km radius is a lot of reach, even for rural off-grids.

Heefi🃏 (MegaMillions)

They need the far witness to make them financially sound

Would love to see your hotspot account so we can look up the awesome witnesses!

^^ Double. Several other miner owners I know live by bays or on peninsula's or near lakes etc, and even with the clear waterways none of us see over 100Km witnesses, and we have antenna's that range from 10m + on ground level and hillsides.

I mean what about when data transfer is important in rural areas like, package tracking etc? Its hard to incentivize people to get on when nobody is really around them and the only way for someone in a rural area to get their 'ROI' is to go with a high DBI antenna to reach the one other dude cities away

19:26

Like what is the point of restricting the coverage at all? If an antenna hits 200km, so be it

Mijelz

I mean what about when data transfer is important in rural areas like, package tracking etc? Its hard to incentivize people to get on when nobody is really around them and the only way for someone in a rural area to get their 'ROI' is to go with a high DBI antenna to reach the one other dude cities away

There is a very little chance that the signal of the sensor with a tiny antenna and reduced tx power would reach to 200km anyway (edited)

1

As the person who probably is most affected by this HIP, and loudest yeller when the 50km chainvar showed up, I'm fine with this version. 100km is a large enough distance and I'm happy we're doing this as a HIP now rather than an unannounced chain var like last time.

1

100km is a good start, but it won't make much difference in small countries or two nearby dense cities.

01:29

I think the distance will be reduced further sooner or later, so why not tie the distance to the transmit scale from now! Transmit scale 1.00 = 100km, 0.99 = 99km.... 0.01 = 1km (edited)

1

Nian

I think the distance will be reduced further sooner or later, so why not tie the distance to the transmit scale from now! Transmit scale 1.00 = 100km, 0.99 = 99km.... 0.01 = 1km (edited)

Not a good ideia. A hotspot with a low transmit scale can still be providing coverage many km away, so they should still be rewarded for it

1

ricopt5

Not a good ideia. A hotspot with a low transmit scale can still be providing coverage many km away, so they should still be rewarded for it

I know it can provide coverage (if high enough), but Transmite scale (reward scale) was introduced for a reason, and by Explorer map, it's not working as it supposed to. Someone with 0.10ts will still have 10km of coverage, which is enough for a such dense area.

@nian it should be increased , countries like Australia the transmit strength limit here is 36DB /4Watts it is significantly stronger than other areas and those distances can be reached readily

MoreDPS

@nian it should be increased , countries like Australia the transmit strength limit here is 36DB /4Watts it is significantly stronger than other areas and those distances can be reached readily

Australia is 4000km, you'll still need more hotspots and off-grids to cover the needed areas. So why not distribute more Hs from now!

Population density buddy. Nobody lives in the desert areas that make a large part of Australia

How large are the batteries on those sensors transmitting at 4W?

Nian

I think the distance will be reduced further sooner or later, so why not tie the distance to the transmit scale from now! Transmit scale 1.00 = 100km, 0.99 = 99km.... 0.01 = 1km (edited)

Bad idea, 150 km max on all and good

MoreDPS

Population density buddy. Nobody lives in the desert areas that make a large part of Australia

I know, but you mention Australia as a whole. Populated areas are already covered as I can see!

The op of this Hip hasn't addressed that different regions have different transmit restrictions and allowable levels

02:04

To set one distance for all when areas can transmit 2 ormore times further would be pointless

secretboy

Bad idea, 150 km max on all and good

The limit they proposed is 100km, and you can vote no with or without my suggestion/idea!

Not going to debate distance this whole idea is baked if they don't do it by legally allowed maximum transmit strength by region

Nian

I know it can provide coverage (if high enough), but Transmite scale (reward scale) was introduced for a reason, and by Explorer map, it's not working as it supposed to. Someone with 0.10ts will still have 10km of coverage, which is enough for a such dense area.

You're missing the point. Let's give an extreme example. Imagine a hex has a hotspot in a basement providing almost no coverage and another hotspot in a tower. Both hotspots will be limited to 50km, which is very unfair to the tower one because it is the only one providing useful coverage. The witnessing distance limit is something that should be tied to sensor capabilities so that PoC behaves more like real network usage, not tied to PoC itself. (edited)

ricopt5

You're missing the point. Let's give an extreme example. Imagine a hex has a hotspot in a basement providing almost no coverage and another hotspot in a tower. Both hotspots will be limited to 50km, which is very unfair to the tower one because it is the only one providing useful coverage. The witnessing distance limit is something that should be tied to sensor capabilities so that PoC behaves more like real network usage, not tied to PoC itself. (edited)

Why would someone in a basement or ground floor keep his miner ON or where it is if his not gonna earn anything!

Nian

Why would someone in a basement or ground floor keep his miner ON or where it is if his not gonna earn anything!

I said "extreme example"

50km can still land you 2HNT/day

1

I find it really enjoyable that so many people join these kind of HIP's to push their own narrative. I think the math in the HIP is pretty solid, as I've verified it myself after having questions about the 10k sample size. Anyone wanting to 'bikeshed', as hashc0de said it, another number should at least provide a similar analysis.

1

backtran

Make it 8.3km. None of my witnesses are further away than that.

It's time to unbox your miner now - from it's location under the bed in the basement

1

Tango

It's time to unbox your miner now - from it's location under the bed in the basement

I will first have to dig it up.

I've personally have never seen a device go over 100km. I've read an article of someone doing it with a very custom setup. I've seen evidence of hotspots but at this stage with helium. People slap on massive antennas and get them 50+ meters above the ground. PoC is meant to show coverage for devices. It's pointless if you have a tower setup that's able to beacon a sister tower 100km when a device in the middle can't see either one. (edited)

3

Hi everyone when the light hotspot hits in the system with the update our full hotpot’s become light spots as well ?

TAha

Hi everyone when the light hotspot hits in the system with the update our full hotpot’s become light spots as well ?

19.04/20.04, see #announcements for more information

100 km (I.e., 62 miles) of coverage per setup is sufficient, IMO. I see nothing wrong with setting a reasonable limit.

1

Rolling distance limit. Sensor-TOU. Daytime (100) sensorpower batt+solar , Nighttime (50) sensorpower batt. Power efficiency is key here.

While the 100km is not that bad it still kills a good part of the action for good offgrid setups, you can provide really good coverage even over 100km when you are at 1500m elevation. But what's terrible here is Helium's recurring way of solving problems by putting band-aids on them.

1

10:16

yeah but Helium team has the wrong mindset, only thinking of the easiest and most convenient solution to them, not about what's logical and fair.

1

10:21

But these are bad decisions long term and they erode people's trust in the network. You invest a lot of time and money in good setups only to find out the next hip will cut in half your rewards. I know for sure i'm not investing more, it's like they are trying to make everybody earn equally, not proportional to the coverage they offer. And i'm not personally affected by this, i don't have any offgrids, i just don't like their approach.

1

Fuck helium

1